US7226595B2

US7226595B2 - Modified Chimeric superantigens and their use

Info

Publication number: US7226595B2
Application number: US10/283,838
Authority: US
Inventors: Per Antonsson; Per Bjork; Mikael Dohlsten; Terje Kalland; Lars Abrahmsen; Johan Hansson; Göran Forsberg
Original assignee: Active Biotech AB
Current assignee: Active Biotech AB; Pfizer Health AB
Priority date: 1996-03-29
Filing date: 2002-10-30
Publication date: 2007-06-05
Also published as: US20030092894A1; HUP9903444A2; ATE294818T1; AU2525197A; CZ371297A3; IL122211A; ID16512A; TW517061B; PL323626A1; JP4114951B2; AU707827B2; EP0835266B1; WO1997036932A1; CA2222757C; CA2222757A1; EP0835266A1; NO975435D0; BR9702179A; NZ329145A; DE69733179D1

Abstract

A conjugate between a target-seeking moiety and a modified superantigen, characterized in that the superantigen is a wild-type superantigen (SA I) in which an amino acid residue in a superantigen region (region I) determining binding to TCR, preferably TCRVβ, and T cell activation have been replaced by another amino acid residue while retaining the ability to activate a subset of T cells.

In preferred embodiment the modified superantigen is a chimer between at least two wild-type superantigens (SA I, SA II etc) characterized in that one or more amino acid residues in a region determining binding to TCR and T cell activation have been interchanged between various wild-type superantigens.

A therapeutic method making use of modified/chimeric superantigens as defined in the preceding paragraphs.

An antibody preparation in which the cysteine residues that provide for interchain disulfide bonds have been mutated so as to forbid interchain disulfide bridges, preferably to serine residues, for use as pharmaceutical.

Description

This application is a divisional application of U.S. application Ser. No. 08/695,692, now U.S. Pat. No. 6,514,498, which was filed on Aug. 12, 1996 and claims priority from Swedish Patent Application No. 9601245-5, which was filed Mar. 29, 1996, and is incorporated herein by reference.

FIELD OF INVENTION

The present invention relates to functionally active modified superantigens which are wild-type superantigens (SA I) in which one or more amino acid residues have been substituted while maintaining superantigen function. In case one or more of the substituting residues (or a conserved amino acid residue thereof) occur in the corresponding positions in another wild-type superantigen (SA II), the modified superantigen is called a chimera. Chimeric superantigens thus will contain part sequences/regions driving from at least two different wild-type superantigens.

By the term “corresponding” is mean that residues, part sequences and regions replacing each other have functionally the same position in superantigens I and II so that substitution will lead to a chimeric form that is able to function as a superantigen.

The terminology grafted/grafting/graft is used in connection with parts of the full sequence of superantigen II that have replaced corresponding parts of superantigen I, even if only one single amino acid has been replaced.

Modified/chimeric superantigens also encompass functional superantigens modified in other ways, for instance conjugated to a target-seeking moiety, including also fused forms when the moiety is a polypeptide/protein. See below.

Superantigens

According to the very first definition (around 1988–1993), superantigens are bacterial or viral proteins capable of binding to MHC class II antigens without intracellular processing and activate T cells by binding to the β-chain variable region (Vβ) of the T cell receptor (TCR). The binding leads to a Vβ family restricted activation of a relatively large proportion/subset of T cells and lysis of MHC Class II expressing cells (superantigen dependent cell cytolysis=SDCC).

Well known wild-type superantigens according to the definition above are the staphylococcal enterotoxins (SEA, SEB, SEC1, SEC2, SED, SEE and SEH). Further examples are Toxic Shock Syndrome Toxin 1 (TSST-1, also of staphylococcal origin), Exfoliating Toxins (EXft), Streptococcal Pyrogenic Exotoxin A, B and C (SPE A, B and C), Mouse Mammary Tumor Virus proteins (MMTV), Streptococcal M proteins, Clostridial Perfringens Enterotoxin (CPET), mycoplasma arthritis superantigens etc. For a review of superantigens and their properties see Kotzin et al 1993.

During the early nineties it was discovered that activation and subsequent cell lysis could occur in a MHC class II independent manner in case the superantigen was conjugated with a target-seeking moiety capable of binding to a cell surface structure (Dohlsten et al WO9201470 and Abrahmsén et al WO9601650). Upon incubation of target cells (carrying the target structure for the target-seeking moiety) and effector cells (T cells) with the conjugates, the target cells become lysed (superantigen antibody dependent cell cytolysis=SADCC) without any requirement for class II expression. Accordingly the superantigen concept of today and used in the context of the present invention, if not otherwise specified, encompasses any compound (preferably of polypeptide structure) that is capable of binding to a cell surface structure (target structure) and to one or more polymorphic TCR chain, in particular the Vβ chain, thereby activating a subset of T cells expressing the specific TCR chain involved in the binding. The T cells then become cytotoxic for cells carrying the surface structure (target structure, target cells). Normally the activated subset of T cells constitutes about 1–20% of the total amount of T cells of an individual.

Background Art—Structural and Functional Studies Utilizing Mutated and Chimeric Superantigens

Chimeric superantigens including point mutated forms have previously been described (Kappler et al WO 9314364, Kappler et al 1992; Grossman et al 1991; Hufnagle et al 1991; Hartwig et al 1993; Fraser et al 1993; Mollick et al 1993; Erwin et al 1992; and Hudson et al 1993). Mollick et al and Hudson et al show from studies of chimeras that the Vβ specificity of SEA and SEE resides in certain amino acid sequences present in the carboxy terminal region (i.e. amino acid residues 200, 206 and 207). In addition to the Vβ specificity, mainly depending on this region, Mollick et al also were able to show that for complete reconstitution of SEE like activity of SEA containing SEE grafts towards Vβ8, a fragment containing the N-terminal 70 amino acid residues from SEE was needed. This fragment contains parts of the SEE-like MHC class II α chain binding site and chimeric SEA/SEE molecules containing this part from SEE, inhibited binding of SEA to MHC class II DR1 in a SEE-like manner.

Recently SEE-SEA chimeras involving an exchange of regions involved in binding to TCRVβ have been described (Lamphaer et al., J. Immunol. 156 (Mar. 15, 1996) 2178–2185). A SEE superantigen Fab antibody fusion protein in which the SEE domains involved in the interaction with T cells have been replaced with the corresponding non-homologous SEA domains has been discussed at ABRF'96: Biomolecular Techniques, Holiday Inn Golden Gateway, San Francisco, Calif. Mar. 30–Apr. 2, 1996 (Björk et al., M45).

Background Art—Therapeutic Use of Superantigens

Non-conjugated superantigens have been suggested for therapy with curative effective presumably being accomplished through a general activation of the immune system (Kalland et al WO9104053; Terman et al WO9110680 and WO9324136; Newall et al 1991).

It has also been suggested to use modified superantigens conjugated to target-seeking moieties (Dohlsten et al WO9201470; Abrahmsén et al WO9601650, both hereby being incorporated by reference). This enabled a broader therapeutic use of T cell activation through Vβ. The conjugates studied so far have had a diminished class II affinity, which in turn has lead to a decrease of the severe systemic toxicity normally associated with the wild-type superantigens.

Terman et al (WO9110680; WO9324136) in side-sentences suggested cancer therapy with modified superantigens and superantigen fragments.

Kappler et al (WO9314634) have suggested to use non-conjugated superantigens mutated to have lost their Vβ-binding ability (in the context of vaccines). Abrahmsén et al (WO9601650) have suggested cancer therapy with conjugated superantigens having a modified, preferably decreased, ability to bind to Class II antigens. The modifications encompassed single mutations as well as construction of chimeras between different superantigens.

The Problems that have Been the Objective to Solve with the Present Invention.

The sera of human populations normally contain high titers of antibodies against superantigens. For the staphylococcal superantigens, for instance, the relative titers are TSST-1>SEB>SEC1>SE3>SEC2>SEA>SED>SEE. These relative titers indicate immunogenicity problems and problems with neutralizing antibodies in case SEs are administered parenterally. Based solely on these problems, SEE should be the preferred staphylococcal superantigen. In the context of work with fusion proteins, however, we have found that the ability for T cell MHC class II independent cytotoxicity, superantigen-antibody dependent cell cytotoxicity (SADCC), of SEE conjugates is poor. The anti-SE titers also indicate that there might be advantages in modifying a “high titer” superantigen to be more like a “low titer” superantigen.

BRIEF SUMMARY OF THE INVENTION

A first objective is to improve the previously known superantigens with respect to lowering their immunogenicity and reaction with neutralizing antibodies.

A second objective is to provide superantigens with less side effects when used as a drug.

A third objective is to provide improved superantigens that can be used as the active principle in the treatment of mammals suffering from cancers, autoimmune diseases, parasitic infestations, viral infections or other diseases associated with cells that on their surface express MHC class II antigens and/or structures that are specific for respective disease and bind to a target-seeking moiety incorporated into the superantigen.

The Discovery that has Resulted in the Invention.

A sequence homology analysis of SEA and SEE (FIG. 2) reveals that the non-identical amino acid residues are concentrated to eight distinct regions. These regions are identified by A, B, C, D, E, F, G, and H as depicted in FIG. 2. For SEA, and SEE the sequences in these regions are identified as follows:


Region	SEQ ID NO for SEA	SEQ ID NO for SEE

A	SEQ ID NO.: 9	SEQ ID NO.: 10
B	SEQ ID NO.: 11	SEQ ID NO.: 12
C	SEQ ID NO.: 13	SEQ ID NO.: 14
D	SEQ ID NO.: 15	SEQ ID NO.: 16
E	SEQ ID NO.: 17	SEQ ID NO.: 18
F	SEQ ID NO.: 19	SEQ ID NO.: 20
G	SEQ ID NO.: 21	SEQ ID NO.: 22
H	SEQ ID NO.: 23	SEQ ID NO.: 24

Outside these eight regions, making up 34% of the sequence, the identity of the two SEs is 97%, with conserved amino acid substitutions accounting for the remaining differences. Four of these regions are structurally close to the two MHC class II binding sites (B: AA 37–50 (SEQ ID NOs. 11 and 12), D: 71–78 (SEQ ID NOs. 15 and 16), E: 136–149 (SEQ ID NOs. 17 and 18), and G 189–195 (SEQ ID NOs. 21 and 22)), and are most likely to interact with the TCR. The additional four regions (A: AA 20–27 (SEQ ID NOs. 9 and 10), C: 60–62 (SEQ ID NOs. 13 and 14), F: 161–176 (SEQ ID NOs. 19 and 20) and H:200–207 (SEQ ID NOs. 23 and 24)) are located on the edge of the molecule, in the vicinity of the putative TCR binding site, postulated to reside in the groove between the two subdomains. By grafting the individual regions (replacement of amino acid residues that differ), we have now found that the property of SEA-conjugates to induce a cytoxic response as well as potentiating proliferative response in the absence of MHC class II, resides in one region tin the TCR binding domain of SEA. This Region (A) is transferable to SEE and have a great impact on activity in the absence of Class II, although limited effects on the Vβ specificity of the superantigen (FIG. 6, Tab 2). All of the regions (A, C, F and H) seem to participate directly or indirectly, in the interaction with the TCR manifested by an altered stimulatory effect on murine T-cell hybridomas (Tab 2).

Due to the analogous mode of action it is conceivable that a similar structural separation of these TCRVβ binding properties is at hand also for superantigens analogous to SEA and SEE. The same may also apply within other types of superantigens, in which the binding structures are organised differently. Our discovery has enabled us to outline the construction of chimeric superantigens that potentially are of extremely great value as therapeutic agents.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1. FIGS. 1A and 1B show Superantigen dependent cellular cytotoxicity (SDCC) and FIG. 1C and FIG. 1D show Superantigen antibody cellular cytotoxicity (SADCC) with C215Fab-SEA and C215Fab-SEE as effector molecules. Cytotoxicity was analyzed in 51Cr release assay using a SEE-reactive human T-cell line (FIG. 1A and FIG. 1C) and a Raji cell line as target and a SEA-reactive human T-cell line (FIG. 1B and FIG. 1D). Target cell lines were Raji (FIG. 1A and FIG. 1B) and Colo 205 (FIG. 1C and FIG. 1D).

FIG. 2. Homology alignment of SEA and SEE. SEA/SEE variable regions close to the TCR binding site (A, C, F and H) and variable regions close to the two MHC class II binding sites.

FIG. 3. Molscript model (Kraulis, 1991) of the SEA crystal (Schad et al. 1995). SEA/SEE variable regions close to the TCR binding site (A, C, F and H) and variable regions close to the two MHC class II binding sites. The zinc ion is a round ball.

FIG. 4. Schematic representation of chimeric SE molecules. Stretches of SEA sequence are depressed. SEA/SEE variable regions are represented by A, B, C, D, E, F, G and H.

FIG. 5. FIG. 5A shows Superantigen dependent cellular cytotoxicity (SDCC) and FIG. 5B shows Superantigen antibody cellular cytotoxicity (SADCC) of C215Fab-SEE/A-A, C215Fab-SEE/A-C, C215FAB-SEE/A-F, C215Fab-SEE/A-H, C215Fab-SEE/A-AH and C215Fab-SEA/E-BDEG. Cytotoxicity was analyzed in a 5 1Cr release assay using a SEA-reactive human T-cell line and Raji (FIG. 5A) or Colo 205 (FIG. 5B) cell lines as targets.

FIG. 6. FIG. 6A shows Superantigen dependent cellular cytotoxicity (SDCC) and FIG. 6B shows Superantigen antibody cellular cytotoxicity (SADCC) with C215Fab-SEA, C215Fab-SEE, C215Fab-SEE/A-A, C215Fab-SEE/A-C, C215Fab-SEE/A-F, C215Fab-SEE/A-H, C215Fab-SEE/A-AH and C215Fab-SEA/E-BDEG as effector molecules. Cytotoxicity was analyzed in a 51Cr release assay using a Vβ22 selected SEA-reactive human T-cell line and Raji (FIG. 6A) or Colo 205 (FIG. 6B) cell lines as targets.

FIG. 7. Seroreactivity in a human Ig pool (Pool of >5000 sera from healthy donors in Southern Europe against C215Fab-SE fusion proteins. Serially diluted human Ig was allowed to interact for 1 h at room temperature with C215Fab-SEA wt, C215Fab-SEE wt, C215Fab-SEE/A-A, C215Fab-SEE/A-H and FabSEE/A-AH immobilized to the micro titer plates at a concentration of 1 ng well.

DETAILED DESCRIPTION OF THE INVENTION

The first aspect of the invention is a method for the treatment of a disease in a mammal by activation of its immune system through administration of a therapeutically effective (immune activating) amount of a modified, preferably chimeric, superantigen. The mammal is preferably a human. The diseases in question are mostly associated with cells expressing on their surface a target structure binding to the superantigen. The target structure is in most cases different from the TCR epitope normally binding to superantigens. Binding to the target structure permits also binding to TCR and T cell activation. Illustrative examples are MHC class II antigens and other cell surface structures that may be expressed on cells associated with the courses of diseases. Illustrative diseases are cancers (such as carcinoma, sarcoma, and melanoma), viral infections, parasitic infestations and autoimmune diseases. The cells expressing the target structure may also be cells that in some way control the development of the disease to be treated.

The characteristic feature of the method is that one employs a modified superantigen in which one or more amino acid residue in a region (region I) providing for binding to a subset of T cells via a polymorphic TCR chain, in particular TCRVβ, in a wild-type superantigen (SA I) has been replaced with a respective amino acid residue retaining superantigen activity to the so modified superantigen. The presently preferred embodiments refer to a chimeric superantigen in which one or more amino acid residue in a region (region I) of a first wild-type superantigen (SA I) have been replaced with the corresponding one or more amino acid residues in a corresponding region (region II) of a second wild-type superantigen (SA II). The regions I and II differ with respect to amino acid sequences. The superantigens I and II have been selected so that the regions I and II can replace each other without killing the superantigen function. In this context one has to account for the fact that a certain region I alone may not be interchangeable with the corresponding region of another wild-type superantigen although when interchanged together with other regions determining TCR binding and T cell activation, the result becomes a functional active superantigen. The regions concerned normally comprise less than 20 residues, in particular for superantigens analogous to SEA. The replacing amino acid residue thus is different from the replaced residue, and conceivably includes also conserved substitutions and other amino acid substitutions leading to functionally active modified superantigens allowing binding to TCRVβ and activation of a subset of T cells. This means that the inventively modified superantigens in which one or more amino acids in the aforementioned regions have been functionally replaced.

The term “conserved substitution” refers to replacement of an amino acid residue by a chemically similar residue, e.g., a hydrophobic residue for a separate hydrophobic residue, a charged residue for a separate charged residue etc.

As superantigens I, II etc., the staphylococcal enterotoxins, in particular those that coordinate zinc, were at the priority date preferred, i.e. SEA, SEE, SED and possibly also SEH.

The regions involved may have either of the above-mentioned functions (se the heading “The Discovery that has resulted in the Invention” and the Experimental Part):

- 1. A great impact on the superantigen activity as such and a limited effect on the TCR specificity, in particular on Vβ specificity. For SEA-type superantigens this means region A (SEQ ID NO. 9) (amino acid positions 20–27).
- 2. A profound effect on the specificity with respect to binding to polymorphic TCR chains, such as the Vβ chain. For SEA-type of superantigens this means regions C (SEQ ID NO. 13) (amino acid positions 60–62), F (SEQ ID NO. 19) (amino acid positions 110–126) and H (SEQ ID NO. 23) (amino acid position 200–207).

For SEA-like superantigens this means one or more of the substitutions (applied to grafting from SEA to SEE; SEE/A chimeras):


Region A:	R20G, N21T, S24G, R27K
Region C:	G60D, P62S
Region F:	H111R, H114Q, G115Y, F117Y, G118N, S124V, G126D
Region H:	D200G, P206S, D207N

At the priority date it was preferred to carry out all substitutions for each region. For other superantigens, analogous substitutions between corresponding positions/regions could conceivable also be carried out.

Typically one could start from one first superantigen, like SEE and SED, and then replace one or more of its unique Vβ binding regions with the corresponding region(s) of a second superantigen (e.g. SEA), the first and second superantigens preferably being selected so that the antibody titer in normal human sera for the first superantigen is lower than for the second superantigen. For SEA and SEE chimeras, the best modes correspond to the chimeras SEE/A-A, SEE/A-AH, and SEA/E-BDEG, with absolute preference for SEE/A-A. See the experimental part and the figures.

Together with the regions A, C, F and H also amino acid residues at other parts can be exchanged. One type of exchange is to reduce the class II binding ability, because this property is associated with common side effects encountered in superantigen therapy (general immune activation with concomitant systemic release of tumor necrosis factor (TNF) and interferon-γ). For superantigens such as SEA, SED and SEE, positions that are important for the ability to coordinate zinc ions may preferably be changed, i.e. positions 225 and 227, for instance in SEA mutation H225A and in particular D227A will have a positive impact on reducing toxic side effects (see Abrahmsén et al WO9601650 and Fraser et al 1993).

Other substitutions may be performed althroughout the molecule as long as they do not destroy the superantigen function, for instance conserved substitutions, in particular outside regions involved in the binding to class II and TCR. A change in the DNA sequence for altering the MHC class II binding or any other change on the DNA level may be carried out either before or after the change in regions providing for binding to TCR. These other types of modifications can equally well have been introduced prior to the amino acid replacement in Region I. In the context of the present invention, the concept of using a “wild-type superantigen” as start of the modification according to the claim thus primarily refers to the wild-type amino acid sequence in region I outside of which prior modifications may have taken place.

Constructions of chimeric and mutated superantigens can be carried out according to techniques well-known in the art. The switch from a region specific for one superantigen to the corresponding region in another superantigen is done on the genomic level and may be accomplished by replacing a complete sequence or by point mutations of those specific bases that are required to end up in the desired amino acid sequence. See for instance the experimental part and also the prior art references cited above. The term “mutation” comprises replacing, inserting or removing one or more amino acid residues by modifying the DNA sequence coding for the protein to be mutated.

The superantigen to be used in the inventive method can be a non-conjugated superantigen modified as described above, i.e., a modified superantigen lacking a specifically attached target-seeking moiety but with a pronounced ability to bind to both MHC class II antigens and a subset of T cells via TCR. More preferably the modified superantigen, preferably a chimeric superantigen, is conjugated to a target-seeking moiety. In the latter case the preferred variants are fusions between the target-seeking moiety and the modified superantigen. The conjugates as such are novel and are a separate aspect of the invention.

The structures of the inventive conjugates are analogous to earlier known antibody-superantigen conjugates (Dohlsten et al WO9201470; Abrahmsén et al WO9601650, both publications hereby being incorporated by reference), i.e. the conjugates often comply with the formula:
T-B-SA(m)

where T represents the target-seeking moiety, SA(m) the modified, preferably chimeric, superantigen as defined above, and B is a covalent bridge linking T and SA(m) together. T may in principle contain further superantigen moieties (SA(m)), and SA(m) further target-seeking moieties, although in the preferred conjugates there are only one target-seeking moiety and one modified superantigen moiety as defined above.

T can in principle be any structure that is able to bind to a cell surface structure, preferably a disease specific structure. The structure against which T is directed is usually different. The structure against which T is directed is usually different from (a) the Vβ chain epitope to which SA(m) binds, and (b) the MHC class II epitopes to which superantigens bind. The target-seeking moiety is primarily selected among interleukins (e.g. interleukin-2), hormones, antibodies including antigen binding fragments of antibodies, growth factors etc. See for instance Woodworth, Preclinical and Clinical development of Cytokine toxins presented at the conference “Molecular approaches to cancer Immunotherapy”, Ashville, N.C., Nov. 7–11, 1993.

At the priority date, it was preferred that T was an antibody (Fab, F(ab)₂, Fv, single chain antibody etc), with particular emphasis for antibody active fragments (such as Fab), directed towards the so called C242 epitope (Lindholm et al., WO9301301) or more preferably towards the binding epitope for the lung cancer specific 5T4 antibody (Stern et al., WO8907947). This, however, does not exclude that other cancer specific antibodies may function equally well or even better. The term “antibodies comprises monoclonal as well as polyclonal variants, with preference for monoclonal preparations.

T may also be directed towards unique structures on more or less healthy cells that regulate or control the development of a disease.

The bridge B may be selected as previously described (Dohlsten et al WO2901470; and Abrahmsén et al WO9601650), i.e. B shall preferably be hydrophilic and exhibit one or more structure(s) selected among amide, thioether, disulphide, etc. The most prominent bridges are those obtained by recombinant techniques, i.e. the conjugation takes place at the genomic level. In such cases oligopeptide bridges containing hydrophilic amino acid residues, such as Gln, Ser, Gly, Glu, Pro, His and Arg are preferred. Particularly preferred Bs are peptide bridges consisting of 1–10 amino acid residues, with absolute preferences for 3–7 amino acid residues. A typical bridge is the tripeptide GlyGlyPro, SEQ ID NO. 1.

The manufacture of the novel inventive conjugates may be carried out in principle according to two main routes: 1. Recombinant techniques and 2. Chemical linking of a target-seeking moiety T to a modified, preferably chimeric, superantigen (SA(m)) as defined above. These methods are well recognized for the ordinary skilled worker and comprise a large number of variants.

Chemical linking of a modified non-conjugated superantigen to a target-seeking moiety T often utilizes functional groups (e.g. primary amino groups or carboxy groups) that are present in many positions in the compounds. It follows that the final product will contain a mixture of conjugate molecules differing in linking positions, as well as hetero- and homo-conjugates.

For recombinant conjugates (fusion proteins) the obtained conjugate substance will be uniform with respect to the linking position. Either the amino terminal of the chimeric superantigen is linked to the carboxy terminal of the target-seeking moiety or vice versa. For antibodies, such as intact antibodies and antigen-binding fragments (Fab, Fv, single chain antibodies etc), either the light or the heavy chain may be utilized for fusion. At present time recombinant conjugates are preferred, with utmost preference for Fab fragments and linking of the amino terminal of the chimeric superantigen to the first constant domain of the heavy antibody chain (CH1), without exclusion of the analogous linking to the light chain or to the VH and VL domain that also may give quite good results.

The main host cell for large scale recombinant production of the inventive modified superantigens (fused forms as well as non-conjugated forms) is E. coli. This host provides for in principle two routes: intracellular production and secretion. The latter variant is preferred because it offers purification of correctly folded proteins from the periplasma and from the culture medium. The above does not exclude that it is possible to produce active conjugates also in other host cells, e.g. eukaryotic cells, such as yeast or mammalian cells.

Pharmaceutical Compositions, Dosage and Routes of Administration.

A third aspect of the instant invention is pharmaceutical compositions containing the inventive modified, preferably chimeric, superantigens as defined above (both conjugated and non-conjugated forms). The compositions contemplated are known in the field, except that now they contain the instant inventive superantigen. Thus, the compositions may be in the form of a lyophilized particulate material, a sterile or aseptically produced solution, a tablet, an ampoule etc. Vehicles such as water (preferably buffered to a physiologically acceptable pH value by for instance PBS) or other inert solid or liquid material may be present. In general terms the compositions are prepared by the conjugate being mixed with, dissolved in, bound to, or otherwise combined with one or more water-soluble or water-insoluble aqueous or non-aqueous vehicles, if necessary together with suitable additives and adjuvants. It is imperative that the vehicles and conditions must not adversely affect the activity of the modified superantigen.

Normally the inventive superantigen will be sold and administered in predispensed dosages, each one containing an effective amount of the conjugate that, based on the result now presented, is believed to be within the range of 10 ng–50 mg, such as within 10 ng–1 mg or within 10 μg–50 mg. The exact dosage will vary from case to case depending on the patient's weight and age, route of administration, type of disease, target-seeking moiety, superantigen, linkage (-B-) etc.

The administration routes will be those commonly contemplated within the field, i.e., a target cell killing effective amount or therapeutically active amount of a superantigen modified according to the invention is brought into contact with the target cells. For the indications specified above this mostly means parenteral administration, such as injection or infusion (subcutaneously, intravenously, intraarterial, intramuscularly, intraperitoneal) to a mammal, such as a human being. The modified, preferably chimeric, superantigens contemplated may be administered locally or systemically.

By the term “target killing effective amount” is contemplated that the amount is effective in activating and directing T cells to destroy target cells.

The preferred administration route at the priority date is the same as contemplated for the superantigen conjugates according to Dohlsten et al WO9201470 and Abrahmsén et al WO9601650. This means 1–5 hours intravenous infusion (preferably 4 hours) per day combined with a fever-reducing agent (paracetamol). The administration is to be repeated during some days, for instance 4 days, with care consideration taken for the risk of boostring antibodies directed towards the conjugate.

The inventive superantigens may be administered either as the main therapy or in preferred modes as adjuvant therapy in connection with surgery or other drugs.

In the context of therapy we have found that antibody preparations that are pure with respect to non-covalently associated heavy and light antibody chains provide advantages over preparations that contain antibodies in which the chains are linked together via cystine linkages. Accordingly a fourth aspect of the invention is the therapeutic use of an antibody preparation, in particular an Fab preparation, in which the cysteine residues linking the chains together have been replaced by an amino acid not permitting disulfide formation, for instance serine. The most preferred antibody specificities for this aspect of the invention were at the priority date the C242 mab (Lindholm et al., WO9301302) and the 5T4 mab as defined in the references cited above. In the preferred variants one of the antibody chains is fused to a superantigen that is capable of activating a subset of T cells in a Vβ specific manner as described above. The superantigen may be a wild type, a chimer, or a point-mutated version (and combination thereof) as described above or by Dohlsten et al WO9201470 or by Abrahmsén et al WO9601650. This aspect of the invention also comprises pharmaceutical compositions as described above, but containing an antibody preparation as defined for this aspect of the invention instead of a chimeric superantigen.

At the priority date, it was preferred to use the Fab fragment 5T4 antibody (Stern et al, WO8907947) in combination with the SEE/A-A chimer with the mutation D227A. The preferred Fab fragment was mutated in both chains in the position providing interchain disulfide linkage (cys to ser). In order to increase the yield of the antibody/fusion protein when produced in E coli, mutations were also carried out in the Vkappa chain at certain position. See the experimental part.

Materials and Methods

Construction of SEA/SEE Chimeric Genes

Construction of SEA/SEE chimeras were made using the polymerase chain reaction (PCR) based method, sequence overlap extension (Horton et al). PCR reactions were performed with UlTma (Perkin-Elmer) according to manufactures recommendations. PCR produced fragments were cloned in PCR-script (Stratagene, USA) and sequenced to verify the correct sequence. The chimeric superantigen genes were then subcloned in the expression vector pKP889 (Abrahmsén et al 1995), fusing the SE constructs to the heavy chain portion to the Fab fragment of the murine monoclonal antibody C215. The SEA and SEE recombinant fusion proteins were produced as full length polypeptides in accordance with the consensus sequence for signal peptide cleavage (von Heijne 1986)

Protein Expression and Purification

The Esherichia coli K12 strain UL635 were used for expression of the Fab-SE fusion proteins and the SEA mutants as described earlier (Abrahmsén et al 1995). Fab-SE fusion protein were harvested by centrifugation at 5000 g and the supernatant fraction were subjected to purification on protein G Sepharose (Pharmacia Biotech AB, Uppsala, Sweden) as earlier described (Abrahmsén et al 1995). The purity of the affinity purified Fab-SE variants were >90% pure when analyzed by SDS-PAGE.

Cells

The human B-cell lymphoma cell line Raji and human colon carcinoma Colo 205 were cultured in complete R-medium (RPMI-1640 supplemented with 10% fetal calf serum (Gibco BRL, Life Technologies, Ltd. Paisley Scotland) 1 mM glutamine; HyClone Europe, Ltd. Cramlington, 5×10⁻⁵M β-mercaptoethanol; ICN Biomedicals INC. Costa Mesa Calif., 0.1 M NaHCO₃; Seromed Biochrome, 1×10⁻²M Hepes buffer; HyClone Europe, Ltd. Cramlington., 0,1 mg/ml gentamycine; Biological Industries Kibbutz Beit Haemek Israel, 1×10⁻³M sodium pyruvate; HyClone Europe, Ltd. Cramlington). CHO cells transfected with human C215 and CD80 molecules were cultivated in complete R-medium supplemented with 0.5 mg/ml Geniticin (G418) Gibco BRL, Life Technologies, Ltd. Paisley Scotland). Peripheral blood mononuclear cells (PBM) were prepared from heparinized blood from normal donors. The cells were isolated by density centrifugation over Ficoll-Paque as previously described (Dohlsten et al 1991). Human T lymphocytes were to homogenicity by positive selection using MiniMACS columns in conjunction with magnetic beads coated with monoclonal antibodies specific for human CD4 and CD8 (Miltenyi Biotec GmbH, Germany) according to the manufacturers specifications. Human SEA and SEE reactive cell lines was generated as previously described (Dohlsten et al 1994). Human TCR Vβ22 expressing cell line was generated from a primary stimulated SEA reactive cell line using positive selective with magnetic Dynabeads (Dynal A.S., Norway) coated with TCR Vβ22 specific monoclonal antibody (Immunotech, France). Enriched cells contained>95% TCR Vβ22⁺T cells as determined by flow cytometry (data not shown). Murine T-cell hybridomas (I1B3, 2B4 and 11.40) were generated as described (Fleury et al 1991).

Cytotoxicity Assay

Cytotoxicity was measured in a standard ⁵¹Cr release assay after 4 or 6 hours as previously described (Dohlsten et al 1991). Human Colo205 or Raji cells were used as target cells. The effector cells, either SEA or SEE reactive human T cell lines or TCR Vβ22 cell lines, were added at an effector to target ratio of 30:1. ⁵¹Cr-labeled target cells were used in the cytotoxicity assays at 2500 cells/200 ml complete medium in V-bottomed microtiter wells. C215Fab-SEA/E hybrids were added at various concentrations as indicated and ⁵¹Cr release was measured in a g-counter. The percentage specific cytotoxicity was calculated as 100×[c.p.m. experimental release−c.p.m. background release)/(c.p.m. total release−c.p.m. background release)].

Lymphocyte Proliferation Assays

To measure proliferation 10⁵human T cell responders were incubated at 37° C. with 10⁴irradiated (20.000 Rad) stimulator cells in 200 ml complete medium in U-shaped 96-well microtitre plates with varying amounts of C215Fab-SEA/E hybrids for 72 hours. Proliferation was estimated by incorporation of [³H]-thymidine as described (Dohlsten et al 1988).

Analysis of Fab-SAg Induced IL-2 Production.

Murine T—T hybridoma cells (10⁵) were incubated in 200 ml complete R-medium with C215Fab-SEA/E chimeric proteins in the presence of 2×10⁴Raji stimulator cells. After 48 hours, supernatants were harvested and analyzed for presence of murine IL-2. Briefly, cytokine content was analyzed using rat anti-mouse cytokine mAb as catcher antibodies. Purified rat anti-mouse IL-2, biotin-labeled rat anti-mouse IL-2, rIL-2 was purchased from PharMingen (San Diego, Calif.). Biotin-labeled anti-cytokine mAb, Vectastain ABC kit (Vector Laboratories, CA) and peroxidase substate kit (Bio-Rad Laboratories, CA) were used for detection of cytokines. The absorbance was determined in a ImmunoReader NJ2000 (InterMed Roskilde, Denmark) at 405 or 450 nm.

Mutation of 5T4 Fab

Construction of a Vector for Expression of 5T4 FAB-SEA in E. coli.

The FV-encoding portions of 5T4 were cloned from the 5T4 hybridoma, obtained from Dr. Peter Stern (Stern et al., WO8907947). In more detail: cDNA was made from the mRNA, regions of the entire variable domains and parts of the signal sequences as well as the first constant domain of the heavy chain and the constant domain of the light chain were amplified by PCR. The oligonucleotides 5Õ-CAATTTTCTTGTCCACCTTGGTGC-3Õ (SEQ ID NO 2) and 5Õ-ACTAGTCGACATGGATGGAGCTITATCATIyTCTT-3Õ (SEQ ID NO. 3) were used for the heavy chain, resulting in a 553 bp fragment, while the oligonucleotides 5Õ-ACTAGTCGACATGGGCITCAAGATGGAGTCACAkwyyCwGG-3Õ (SEQ ID NO: 4) and 5Õ-GCGCCGTCTAGAATTAACACTCATTCCTGTTGAA-3Õ (SEQ ID NO: 5) were used for the light chain, yielding a 724 bp fragment. For each chain three separate clones were sequenced and found to be identical. DNA fragments suitable for insertion into the expression vector (ref) were obtained in a second PCR step. In order to assemble a Fab-expression plasmid, the variable regions of 5T4 were fused to sequences coding for constant regions from the murine IgG1/k antibody C242 mab (Lindholm et al). WO9301302). A region coding for a superantigen derived from staphylococcal enterotoxin A (SEA) was fused after the heavy chain. The verified sequence for the Vkappa chain antibody framework for the 5T4 antibody is given in the result.

Mutagenesis of 5T4

Seven amino acid replacements were introduced in the regions coding for the antibody framework. These were Phe10Ser, Thr45Lys, Ile63Ser, Tyr67Ser, Phe73Leu, Thr77Ser and Leu78Val. Similarly, the Cys residue in either chain involved in the interdomain disulfide bond were replaced by serine residues resulting in the mutations Cys458Ser in the heavy chain and Cys214Ser in the light chain. The mutations were introduced using PCR-based mutagenesis and the DNA sequence obtained were confirmed using sequencing.

Fermentor Expression and Purification of 5T4Fab-SEA.

The expression plasmid contains the kanamycin resistance gene and a lacUV5-promoter that may be induced with IPTG. The fusion proteins were purified from the clarified culture medium using protein G Sepharose and SP-Sepharose (Pharmacia Biotec, Uppsala, Sweden) and formulated in citrate buffer using Sephadex G-25, essentially as described. Characterization using SDS-PAGE, reverse phase HPLC and mass spectrometry showed that the purified fusion protein was more than 95% pure and had the correct molecular mass.

Results

Superantigen Modification

The superantigen dependent cellular cytotoxicity (SDCC) of C215Fab-SEA and of C215Fab-SEE against MHC class II⁺Raji cells, were analyzed using SEA- and SEE-reactive human T cells as effector cell lines. Despite the difference in Vβ specificity between SEA and SEE both superantigens exhibited induction of comparable degree of cytotoxicity with both effector cell lines (FIG. 1). To discriminate between effects of MHC class II presentation and direct effects of SEA and SEE in TCR recognition, they were examined in superantigen-antibody dependent cellular cytotoxicity (SADCC) against C215 expressing Colo205 cells. In this assay the Fab moiety directs the fusion protein to C215-expressing target cells and results in the presentation of fused SE molecules to cytotoxic T-cells (CTL) independent of MHC class II molecules (Dohlsten et al 1994). Despite 80% amino acid sequence identify between SEA and SEE the TCR interaction of SEA and SEE displays qualitative differences in this type of assay. The C215Fab-SEA fusion protein retains its ability to direct SEA and SEE reactive CTL against the MHC class II—target cells (FIG. 1) while C215Fab-SEE fails to induce cytotoxicity of the, MHC class II—target cells, neither with SEA nor with SEE reactive CTL (FIG. 1).

It has previously been reported by other investigators that the differences in Vβ specificity between SEA and SEE primarily relates to a three amino acid difference in the loop preceding and in the irregular a5 helix (Irwin et al 1992, Hudson et al 1993, Fraser et al 1993, and Mollick et al 1993). The difference in respect to TCR interaction reported in this investigation is not related to altered TCR Vβ specificity since the ability of C215Fab-SEA to induce MHC class II independent cytotoxicity is not restricted to SEA reactive CTL but is also seen with SEE reactive CTL.

Sequence homology analysis of SEA and SEE (FIG. 2) reveals that the non-identical amino acid residues are concentrated to eight distinct regions. Outside these eight regions, making up to 34% of the sequence, the identify of the two SE's is 97%, with conserved amino acid substitutions accounting for the remaining differences. Four of the non-homologous regions are structurally close to the two MHC class II binding sites (B(SEQ ID NOs. 11 and 12), D (SEQ ID NOs. 15 and 16), E (SEQ ID NOs. 17 and 18) and G (SEQ ID NOs. 21 and 22), and are not likely to interact with the TCR (FIG. 3). The additional four regions (A: AA 20–27 (SEQ ID NOs. 9 and 10), C: 60–62 (SEQ ID NOs. 13 and 14), F: 161–176 (SEQ ID NOs. 19 and 20) and H: 200–207 (SEQ ID NOs. 23 and 24)) are located on the edge of the molecule (FIG. 3), in the vicinity of the TCR binding site, located in the groove between the two subdomains (Kappler et al 1992). To investigate the qualitative difference in TCR recognition between SEA and SEE we made hybrid proteins by grafting the regions from SEA to SEE as single region chimeras (SEE/A-A, -C, -F, H) as double region hybrids (SEE/A-AH) and by grafting the regions located in the vicinity of the MHC class II binding sites on SEE to SEA (SEA/E-BDEG) (FIG. 4). All of the chimeric SEs were expressed as C215Fab fusion proteins to be able to detect differences in respect to their activity in the absence of MHC class II.

The SEA/E Hybrid Proteins in Fusion with the C215Fab Moiety Displays Difference in Fab Targeted Cytotoxic Assays.

The SDCC activity of C215Fab-SEE/A hybrid proteins against MHC class II⁺Raji cells were analyzed using SEA-reactive human T cells as effectors. The EC₅₀values of all C215Fab-SE hybrids as well as the C215Fab-SEAwt and -SEEwt falls in the margin of errors (e.g. 10-¹²-10-¹¹M, FIG. 5). The only detectable difference are slightly reduced plateau of the C215Fab-SEE/A-AH hybrid, indicating a loss of responding T cells. On the other hand in SADCC experiments were the cytotoxicity is directed towards MHC class II-C215⁺Colo 205 cell line, only C215Fab-SEE/A-A, C215Fab-SEE/A-AH and C215Fab-SEA/E-BDEG induced comparable cytotoxicity as the C215Fab-SEAwt (FIG. 5). The C215Fab-SEE/A-F hybrid is able to induce C215 targeted cytotoxicity at higher concentrations EC₅₀>10⁻¹⁰M). Although the C215Fab-SEE/A-H hybrid is able to induce C215 targeted cytotoxicity with similar half maximal concentrations as to C215Fab-SEAwt (e.g. EC₅₀10⁻¹³M) is the absolute level of cytotoxicity strongly reduced (FIG. 5). This difference could be a consequence of a restricted Vβ specificity of the C215Fab-SEE/A-H while the ability of inducing C215 targeted cytotoxicity prevails in the responding T cell sub-population. To further investigate this notion we prepared human Vβ22 oligoclonal CTL line. Human Vβ22 are analogous to murine Vβ3 in the respect that it is a SEA non SEE specific Vβ family. It has previously been shown (Mollick et al 1993) that the major contribution of SEA and SEE Vβ is primarily residing in the three amino acid difference between SEA and SEE in region H (AA 200–207). In SDCC assays against MHC class II⁺Raji targets, using the Vβ22 oligoclonal CTL line as effectors, only hybrids containing the SEA-H region are able to give C215Fab-SEA wt-like response (e.g. C215Fab-SEE/A-H, C215Fab-SEE/-AH and C215Fab-SEA/E-BDEG, FIG. 6). The C215Fab-SEE/A-A hybrid, that were able to induce a full SDCC response with whole CTL population as effectors is in this assay strongly reduced both in half maximal concentration and in the plateau (FIG. 6). When the cytotoxicity of the Vβ22 CTL is directed towards the MHC class II⁻/C215⁺Colo 205 cell line only hybrids containing both SEA-A and SEA-H (e.g. C215Fab-SEE/A-AH and C215Fab-SEA/E-BDEG) regions are able to induce a cytotoxic response, comparable to a C215Fab-SEAwt (FIG. 6). The hybrid containing only the SEA region A (C215Fab-SEE/A-A) induces a lower level of cytotoxicity with a comparable EC50 value. This indicates that the remaining activity seen with the C215Fab-SEE/A-H hybrid in SADCC with the whole T cell population as effectors is not a consequence of the hybrid induced response in restricted population of T cells. A more likely explanation for the observation is that the ability to induce a SADCC response of the C215Fab SE hybrid proteins is primarily residing in the SEA-A region with a minor contribution from the SEA-H and -F regions. There is no evidence that this quality is restricted to any subset of T cells in the combined SEA-SEE responding T cell population, since C215Fab SEA is able to induce the same response with as well with SEE reactive CTL<<s and C215Fab-SEE/A-A is able to fully reconstitute the response seen with C215Fab-SEA.

The SEA/E Hybrid Proteins in Fusion with the C215Fab Moiety Displays Difference in Fab Targeted Proliferation Assays.

It has been previously been shown that purified resting human T cells are induced to proliferate by presentation of C215Fab-SEA on a MHC class II⁻/C215⁺cell line (Lando et al 1993). The ability of C215Fab-SEA to induce MHC II independent proliferation is however markedly reduced with C215Fab-SEE (Tab. 1). To investigate if this difference in quality shows the same confinement to SEA region A, as was seen with SADCC, we investigated the proliferative capacity of C215Fab-SE hybrids, presented by either CHO-DR1⁺/CD80⁺transfected cell lines, on purified resting human T cells. When presenting the Fab-SE conjugates on CHO-DR1⁺/CD80⁺no differences between the different SE proteins were noted (data not shown). However grafts of SEA region A, C and H in SEE potentates the proliferative activity compared to C215Fab-SEE. The best results were obtained by grafting SEA regions A and H, indicating an important role for region A as was seen for the MHC class II independent cytotoxicity. By using a negative selection it is possible that the differences between Fab-SEA and -SEE would be more prominent.

Vβ Specificity of SE-hybrids

To further investigate if the C215Fab-SEA/SEE hybrid-fusion proteins were associated with a certain Vβ specificity we used SEA reactive murine T cell hybridomas expressing Vβ1, Vβ3 and Vβ11. It is obvious from the data obtained that all of the regions investigated, directly or indirectly, affects the interaction with the TCR. By grafting SEA regions C and F in C215Fab-SEE the activity towards the SEA and SEE cross reactive Vβ1 hybridoma I1B3 is destroyed. The same chimeras seem to have no or minor effects on the activity of Vβ3 and Vβ11 hybridomas (2.B4 and 11.40) in comparison with C215Fab-SEE. By grafting SEA region A in C215Fab-SEE the activity towards Vβ3(2.B4) is enhanced by at least a factor 100, in comparison to C215Fab-SEE. More pronounced effects are seen with the same cell line by grafting SEA region H in C215Fab-SEE. This pronounced effect on the influence of Vβ3 specificity by SEA region H has also been noted by earlier investigations (Mollick et al 1993). The same chimera however (C215Fab-SEE/A-H), seem to reduce the activity towards the SEA/SEE cross reactive Vβ1 and Vβ11 hybridomas (I1B3 and 11.40) by a factor 10. In conclusion, the TCR interaction of SEA seems to involve all of the SEA-SEE, variable A, C, F and H.

Seroreactivity

The seroreactivity in human serum samples towards the chimeric SEs was investigated both in pooled samples from different parts of the world as well as in individual serum samples. By grafting both SEA regions A and H in SEE we obtained an intermediate seroreactivity (FIG. 7). A similar seroreactivity was also seen against the chimera C215Fab-SEE/A. However, single grafts of SEA region A in SEE (C215Fab-SEE/A-A) gave a C215Fab-SEE like seroreactivity, indicating that SEA region H is responsible for the remaining seroreactivity against C215Fab-SEE/A-AH. This indicates that the SEA region H is part of dominating antigenic epitope in SEA. The seroreactivity from pooled serum samples from other parts of the world (Japan and USA) as well as 14 individual samples from Sweden all confirms the same general pattern (data not shown).

Mutations of the Fab Part of the Fusion Protein

Expression of 5T4FabSEA-Constructs

The production level in E. coli of 5T4Fab-SEA in the fermenter was found to significantly lower than other Fab-superantigen constructs previously studied in our lab. Two types of modifications were therefore introduced to increase the production level. Firstly, seven different point mutations in the framework region of the light chain were introduced. These were Phe10Ser, Thr45Lys, Ile63Ser, Tyr67Ser, Phe73Leu, Thr77Ser and Leu78Val. Secondly, the cysteine residues making the disulfide bond connecting the heavy and the light chain were replaced by serine residues. The latter modification resulted in a three-fold increase and the 7 point mutations in an additional 12-fold increase in the production level. In addition to the significantly increased production level, removing the disulfide bond also gives a more homogenous product since the possibility of these reactive thiol groups to react with other thiol containing agents is excluded.

The modified 5T4 molecule were checked for affinity in its antigen as well as the biological activity in SADCC. No differences between the mutant form and the wildtype form could be detected in these assays.

The Cys/Ser mutation was also performed in the heavy and light chain of the Fab fragments of several other monoclonal antibodies. The products became homogenous and fully retained the antigen binding capability.

Sequence of region of the antibody frame work for the 5T4 Vkappa chain:

DAVMTQTPTF LLVSAGDRVT ITC KASQSVS	50	(SEQ ID NO 6)
NDVA WY QQ KP G Q SPTLLISY

TSSRYA GVPD RF IGSGYGTD FTFTISTLQA	100
EDLAVFC QQ DYNSPPT FGG GTKLEIK

Underlined sequences are CDRs. Bold-typed positions were mutated: Phe10Ser, Thr45Lys, Ile63Ser, Ile63Thr, Tyr67Ser, Phe73Leu, Thr77Ser, Leu78Val.

	TABLE 1

	Proliferation EC₅₀(pM)

	C215Fab-SEAwt	2.2
	C215Fab-SEEwt	6.9
	C215Fab-SEE/A-A	0.9
	C215Fab-SEE/A-C	2.8
	C215Fab-SEE/A-F	5.7
	C215Fab-See/A-H	1.0
	C215Fab-SEE/A-AH	0.3
	C215Fab-SEA/E-BDEG	1.6

TABLE 2

I1B3	2,B4	11,49
(MuVβ 1)	(MuVβ 3)	(MuVβ 11)
EC₅₀(nM)	EC₅₀(nM)	EC₅₀(nM)

C215Fab-SEA	10	3	0.05
C215Fab-SEE	10	>1000	0.05
C215Fab-SEE/A-A	10	10	0.05
C215Fab-SEE/A-C	>1000	>1000	0.05
C215Fab-SEE/A-F	>300	>300	0.05
C215Fab-SEE/A-H	100	3	0.3
C215Fab-SEE/A-AH	10	3	0.3

Correction for C215Fab binding to serum proteins was made by subtracting the OD-value for C215Fab at each point. Each point represents the mean of duplicate samples. For further details see Materials and methods.

Table 1. Purified human T-cells were stimulated for 96 h with respective C215Fab-SE presented on MHC class II negative CHO-CD80/C215 transfectants. After 72 h the cells were pulsed with 3H-thymidine for 24 h and incorporated label were measured and represented as half maximal concentration (EC50).

Table 2. Murine T cell hybridomas were stimulated for 48 h with native or chimeric Fab conjugated superantigen. Activity was measured as IL-2 production and represented as half maximal concentration (EC50).

Because many varying and different embodiments may be made within the scope of the inventive concept herein taught, and because modifications may be made in the embodiments herein detailed in accordance with the descriptive requirements of the law, it is to be understood that the details herein are to be interpreted as illustrative and not in a limiting sense.

REFERENCES

Abrahmsén L. et al (1995) Characterization of two distinct MHC Class II binding sites in the superantigen staphylococcal enterotoxin A. EMBO J 14:2978–86.
Abrahmsén et al (1996) WO961650 (patent application)
Dohlsten et al (1988) Two subsets of human peripheral blood CD4+ T helper cells differing in the capacity to produce IL-2 and interferon-gamma can be defined by the Leu-18 and UCHL1 monoclonal antibodies. Eur J Immunol 18:1173.
Dohlsten M et al (1994) Monoclonal antibody-superantigen fusion proteins: Tumor specific agents for T cell based tumor therapy. Proc Natl Acad Sci USA 91:8945–49.
Dohlsten M et al (1991) Monoclonal antibody-targeted superantigens: A different class of anti-tumor agents. Proc Natl Acad Sci USA 88:9287–91.
Dohlsten et al (1992) WO920470 (patent application)
Fleury S et al (1991) Mutational analysis of the interaction between CD4 and class II MHC: class II antigens. Cell 66:1037–49.
Fraser J D et al (1993) Structural model of Staphylococcal entertoxin A interactions with MHC class II antigens. In: Huber, B T Palmer, E (eds) Current Communications in Cell and Molecular Biology 7. Cold Spring Harbour Laboratory Press, Cold Spring Harbor, N.Y.
Grossman et al (1991) Mutation of the disulfide loop in staphylococcal enterotoxin A. Consequences for T cell recognition. J Immunol 147:3274–81.
Hartwig U F et al (1993) Mutations affecting MHC class II binding of the
Horton R M et al (1990) Gene splicing by overlap extension: tailor-made genes using the polymerase chain reaction. Biotechniques 8:528–35
Hudson et al (1993) Two adjacent residues in staphylococcal enterotoxins A and E determine T cell receptor V beta specificity. J Exp Med 177:175–84.
Huffiagle W O et al (1991) The carboxyl-terminal region of staphylococcal enterotoxin type A is required for a fully active molecule. Infect Immun 59:2126–34.
Irwin M J et al (1992) Entertoxin residues determining T-cell receptor Vb binding specificity. Nature 359:841–3
Kalland et al (1991) WO9314634 (patent application)
Kappler J W et al (1992) Mutations defining functional regions of the superantigen staphylococcal enterotoxin B. J Exp Med 175:387–96
Kappler et al (1993) WO9314634 (patent application)
Kotzin B L et al (1993) Superantigens and their potential role in human disease. Adv Immunol 54:99–166.
Kraulis P J (1991) MOLSCRIPT: A program to produce both detailed and schematic plots of protein structures. J Appl Cryst 24:946–50.
Lando P A et al (1993) Co-stimulation with B7 and targeted superantigen is required for MHC class II-independent T-cell proliferation but not cytotoxicity. Immunology 80: 236–241.
Lindholm et al (1993) 9301302 (patent application)
Mollick J A et al (1993) Localization of a site on bacterial superantigens that determines T cell receptor beta chain specificity. J Exp Med 177:283–93.
Newall et al (1991) In vivo T-cell activation by staphylococcal enterotoxin B prevents outgrowth of a malignant tumor. Proc Natl Acad Sci USA 88 1074–1078
Schad E M et al (1995) Crystal structure of the superantigen, Staphylococcal enterotoxin type A. EMBO J 14:3292–301.
Stern et al (1989) WO8907947 (patent application)
Terman et al (1991) WO9110680 (patent application)
Terman et al (1989) WO9324136 (patent application)
Von Heijne, G (1986) A new method for predicting signal sequence cleavage sites. Nucleic Acid Res. 14, 1483–90.

Claims

1. A conjugate comprising a bacterial superantigen and an immunoglobulin, wherein the superantigen comprises Staphylococcal enterotoxin E, SEQ ID NO: 8, mutated in the following positions relative to SEQ ID NO: 8, at position 20, a glycine or a conserved substitution of glycine; at position 21, a threonine or a conserved substitution of threonine; at position 24, a glycine or a conserved substitution of glycine; at position 27, a lysine or a conserved substitution of lysine, and at position 225 and/or 227, an alanine or a conserved substitution of alanine, such that the modified superantigen induces increased cytotoxicity relative to unmodified Staphylococcal enterotoxin E, and reduced binding to endogenous antibodies relative to Staphylococcal enterotoxin A.

2. The conjugate of claim 1, wherein the Staphylococcal enterotoxin has a glycine at position 20.

3. The conjugate of claim 1, wherein the Staphylococcal enterotoxin has a threonine at position 21.

4. The conjugate of claim 1, wherein the Staphylococcal enterotoxin has a glycine at position 24.

5. The conjugate of claim 1, wherein the Staphylococcal enterotoxin has a lysine at position 27.

6. The conjugate of claim 1, wherein the immunoglobulin is an antibody or an antigen-binding fragment of an antibody.

7. The conjugate of claim 6, wherein the immunoglobulin is an antibody.

8. The conjugate of claim 1, wherein the immunoglobulin is an antigen-binding fragment of a monoclonal antibody to the 5T4 antigen.

9. The conjugate of claim 8, wherein the Vkappa chain of the antigen-binding fragment of the monoclonal antibody to the 5T4 antigen comprises SEQ ID NO:6 having the following amino acid substitutions in SEQ ID NO:6, a serine or a conserved substitution of serine at position 10, a lysine or a conserved substitution of lysine at position 45; a serine or a conserved substitution of serine at position 63, a serine or a conserved substitution of serine at position 67, a leucine or a conserved substitution of leucine at position 73, a serine or a conserved substitution of serine at position 77, a valine or a conserved substitution of valine at position 78.